bitmap.c 52.7 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/*
 * bitmap.c two-level bitmap (C) Peter T. Breuer (ptb@ot.uc3m.es) 2003
 *
 * bitmap_create  - sets up the bitmap structure
 * bitmap_destroy - destroys the bitmap structure
 *
 * additions, Copyright (C) 2003-2004, Paul Clements, SteelEye Technology, Inc.:
 * - added disk storage for bitmap
 * - changes to allow various bitmap chunk sizes
 */

/*
 * Still to do:
 *
 * flush after percent set rather than just time based. (maybe both).
 */

18
#include <linux/blkdev.h>
19 20 21 22 23 24 25 26 27 28
#include <linux/module.h>
#include <linux/errno.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/timer.h>
#include <linux/sched.h>
#include <linux/list.h>
#include <linux/file.h>
#include <linux/mount.h>
#include <linux/buffer_head.h>
29
#include "md.h"
30
#include "bitmap.h"
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

/* debug macros */

#define DEBUG 0

#if DEBUG
/* these are for debugging purposes only! */

/* define one and only one of these */
#define INJECT_FAULTS_1 0 /* cause bitmap_alloc_page to fail always */
#define INJECT_FAULTS_2 0 /* cause bitmap file to be kicked when first bit set*/
#define INJECT_FAULTS_3 0 /* treat bitmap file as kicked at init time */
#define INJECT_FAULTS_4 0 /* undef */
#define INJECT_FAULTS_5 0 /* undef */
#define INJECT_FAULTS_6 0

/* if these are defined, the driver will fail! debug only */
#define INJECT_FATAL_FAULT_1 0 /* fail kmalloc, causing bitmap_create to fail */
#define INJECT_FATAL_FAULT_2 0 /* undef */
#define INJECT_FATAL_FAULT_3 0 /* undef */
#endif

#ifndef PRINTK
#  if DEBUG > 0
#    define PRINTK(x...) printk(KERN_DEBUG x)
#  else
#    define PRINTK(x...)
#  endif
#endif

61
static inline char *bmname(struct bitmap *bitmap)
62 63 64 65 66 67 68 69 70 71 72
{
	return bitmap->mddev ? mdname(bitmap->mddev) : "mdX";
}

/*
 * just a placeholder - calls kmalloc for bitmap pages
 */
static unsigned char *bitmap_alloc_page(struct bitmap *bitmap)
{
	unsigned char *page;

73
#ifdef INJECT_FAULTS_1
74 75
	page = NULL;
#else
76
	page = kzalloc(PAGE_SIZE, GFP_NOIO);
77 78 79 80
#endif
	if (!page)
		printk("%s: bitmap_alloc_page FAILED\n", bmname(bitmap));
	else
81
		PRINTK("%s: bitmap_alloc_page: allocated page at %p\n",
82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104
			bmname(bitmap), page);
	return page;
}

/*
 * for now just a placeholder -- just calls kfree for bitmap pages
 */
static void bitmap_free_page(struct bitmap *bitmap, unsigned char *page)
{
	PRINTK("%s: bitmap_free_page: free page %p\n", bmname(bitmap), page);
	kfree(page);
}

/*
 * check a page and, if necessary, allocate it (or hijack it if the alloc fails)
 *
 * 1) check to see if this page is allocated, if it's not then try to alloc
 * 2) if the alloc fails, set the page's hijacked flag so we'll use the
 *    page pointer directly as a counter
 *
 * if we find our page, we increment the page's refcount so that it stays
 * allocated while we're using it
 */
105 106
static int bitmap_checkpage(struct bitmap *bitmap,
			    unsigned long page, int create)
107 108
__releases(bitmap->lock)
__acquires(bitmap->lock)
109 110 111 112
{
	unsigned char *mappage;

	if (page >= bitmap->pages) {
113 114 115 116
		/* This can happen if bitmap_start_sync goes beyond
		 * End-of-device while looking for a whole page.
		 * It is harmless.
		 */
117 118 119 120 121 122 123 124 125 126 127 128 129 130
		return -EINVAL;
	}

	if (bitmap->bp[page].hijacked) /* it's hijacked, don't try to alloc */
		return 0;

	if (bitmap->bp[page].map) /* page is already allocated, just return */
		return 0;

	if (!create)
		return -ENOENT;

	/* this page has not been allocated yet */

131 132 133 134 135
	spin_unlock_irq(&bitmap->lock);
	mappage = bitmap_alloc_page(bitmap);
	spin_lock_irq(&bitmap->lock);

	if (mappage == NULL) {
136 137 138 139 140 141
		PRINTK("%s: bitmap map page allocation failed, hijacking\n",
			bmname(bitmap));
		/* failed - set the hijacked flag so that we can use the
		 * pointer as a counter */
		if (!bitmap->bp[page].map)
			bitmap->bp[page].hijacked = 1;
142 143
	} else if (bitmap->bp[page].map ||
		   bitmap->bp[page].hijacked) {
144 145 146
		/* somebody beat us to getting the page */
		bitmap_free_page(bitmap, mappage);
		return 0;
147
	} else {
148

149
		/* no page was in place and we have one, so install it */
150

151 152 153
		bitmap->bp[page].map = mappage;
		bitmap->missing_pages--;
	}
154 155 156 157 158 159
	return 0;
}

/* if page is completely empty, put it back on the free list, or dealloc it */
/* if page was hijacked, unmark the flag so it might get alloced next time */
/* Note: lock should be held when calling this */
160
static void bitmap_checkfree(struct bitmap *bitmap, unsigned long page)
161 162 163 164 165 166 167 168 169 170 171
{
	char *ptr;

	if (bitmap->bp[page].count) /* page is still busy */
		return;

	/* page is no longer in use, it can be released */

	if (bitmap->bp[page].hijacked) { /* page was hijacked, undo this now */
		bitmap->bp[page].hijacked = 0;
		bitmap->bp[page].map = NULL;
172 173 174 175 176 177
	} else {
		/* normal case, free the page */
		ptr = bitmap->bp[page].map;
		bitmap->bp[page].map = NULL;
		bitmap->missing_pages++;
		bitmap_free_page(bitmap, ptr);
178 179 180 181 182 183 184 185 186 187 188
	}
}

/*
 * bitmap file handling - read and write the bitmap file and its superblock
 */

/*
 * basic page I/O operations
 */

189
/* IO operations when bitmap is stored near all superblocks */
190
static struct page *read_sb_page(mddev_t *mddev, loff_t offset,
191 192
				 struct page *page,
				 unsigned long index, int size)
193 194 195 196 197
{
	/* choose a good rdev and read the page from there */

	mdk_rdev_t *rdev;
	sector_t target;
198
	int did_alloc = 0;
199

200
	if (!page) {
201
		page = alloc_page(GFP_KERNEL);
202 203 204 205
		if (!page)
			return ERR_PTR(-ENOMEM);
		did_alloc = 1;
	}
206

207
	list_for_each_entry(rdev, &mddev->disks, same_set) {
208 209
		if (! test_bit(In_sync, &rdev->flags)
		    || test_bit(Faulty, &rdev->flags))
210 211
			continue;

212
		target = rdev->sb_start + offset + index * (PAGE_SIZE/512);
213

214
		if (sync_page_io(rdev->bdev, target,
215
				 roundup(size, bdev_logical_block_size(rdev->bdev)),
216
				 page, READ)) {
217
			page->index = index;
218 219
			attach_page_buffers(page, NULL); /* so that free_buffer will
							  * quietly no-op */
220 221 222
			return page;
		}
	}
223 224
	if (did_alloc)
		put_page(page);
225
	return ERR_PTR(-EIO);
226 227 228

}

229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262
static mdk_rdev_t *next_active_rdev(mdk_rdev_t *rdev, mddev_t *mddev)
{
	/* Iterate the disks of an mddev, using rcu to protect access to the
	 * linked list, and raising the refcount of devices we return to ensure
	 * they don't disappear while in use.
	 * As devices are only added or removed when raid_disk is < 0 and
	 * nr_pending is 0 and In_sync is clear, the entries we return will
	 * still be in the same position on the list when we re-enter
	 * list_for_each_continue_rcu.
	 */
	struct list_head *pos;
	rcu_read_lock();
	if (rdev == NULL)
		/* start at the beginning */
		pos = &mddev->disks;
	else {
		/* release the previous rdev and start from there. */
		rdev_dec_pending(rdev, mddev);
		pos = &rdev->same_set;
	}
	list_for_each_continue_rcu(pos, &mddev->disks) {
		rdev = list_entry(pos, mdk_rdev_t, same_set);
		if (rdev->raid_disk >= 0 &&
		    !test_bit(Faulty, &rdev->flags)) {
			/* this is a usable devices */
			atomic_inc(&rdev->nr_pending);
			rcu_read_unlock();
			return rdev;
		}
	}
	rcu_read_unlock();
	return NULL;
}

263
static int write_sb_page(struct bitmap *bitmap, struct page *page, int wait)
264
{
265
	mdk_rdev_t *rdev = NULL;
266
	mddev_t *mddev = bitmap->mddev;
267

268
	while ((rdev = next_active_rdev(rdev, mddev)) != NULL) {
269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313
		int size = PAGE_SIZE;
		loff_t offset = mddev->bitmap_info.offset;
		if (page->index == bitmap->file_pages-1)
			size = roundup(bitmap->last_page_size,
				       bdev_logical_block_size(rdev->bdev));
		/* Just make sure we aren't corrupting data or
		 * metadata
		 */
		if (mddev->external) {
			/* Bitmap could be anywhere. */
			if (rdev->sb_start + offset + (page->index
						       * (PAGE_SIZE/512))
			    > rdev->data_offset
			    &&
			    rdev->sb_start + offset
			    < (rdev->data_offset + mddev->dev_sectors
			     + (PAGE_SIZE/512)))
				goto bad_alignment;
		} else if (offset < 0) {
			/* DATA  BITMAP METADATA  */
			if (offset
			    + (long)(page->index * (PAGE_SIZE/512))
			    + size/512 > 0)
				/* bitmap runs in to metadata */
				goto bad_alignment;
			if (rdev->data_offset + mddev->dev_sectors
			    > rdev->sb_start + offset)
				/* data runs in to bitmap */
				goto bad_alignment;
		} else if (rdev->sb_start < rdev->data_offset) {
			/* METADATA BITMAP DATA */
			if (rdev->sb_start
			    + offset
			    + page->index*(PAGE_SIZE/512) + size/512
			    > rdev->data_offset)
				/* bitmap runs in to data */
				goto bad_alignment;
		} else {
			/* DATA METADATA BITMAP - no problems */
		}
		md_super_write(mddev, rdev,
			       rdev->sb_start + offset
			       + page->index * (PAGE_SIZE/512),
			       size,
			       page);
314
	}
315 316

	if (wait)
317
		md_super_wait(mddev);
318
	return 0;
319 320 321

 bad_alignment:
	return -EINVAL;
322 323
}

324
static void bitmap_file_kick(struct bitmap *bitmap);
325
/*
326
 * write out a page to a file
327
 */
328
static void write_page(struct bitmap *bitmap, struct page *page, int wait)
329
{
330
	struct buffer_head *bh;
331

332 333 334 335 336
	if (bitmap->file == NULL) {
		switch (write_sb_page(bitmap, page, wait)) {
		case -EINVAL:
			bitmap->flags |= BITMAP_WRITE_ERROR;
		}
337
	} else {
338

339
		bh = page_buffers(page);
340

341 342 343 344 345 346 347
		while (bh && bh->b_blocknr) {
			atomic_inc(&bitmap->pending_writes);
			set_buffer_locked(bh);
			set_buffer_mapped(bh);
			submit_bh(WRITE, bh);
			bh = bh->b_this_page;
		}
348

349
		if (wait)
350 351
			wait_event(bitmap->write_wait,
				   atomic_read(&bitmap->pending_writes)==0);
352
	}
353 354
	if (bitmap->flags & BITMAP_WRITE_ERROR)
		bitmap_file_kick(bitmap);
355 356 357 358 359 360
}

static void end_bitmap_write(struct buffer_head *bh, int uptodate)
{
	struct bitmap *bitmap = bh->b_private;
	unsigned long flags;
361

362 363 364 365
	if (!uptodate) {
		spin_lock_irqsave(&bitmap->lock, flags);
		bitmap->flags |= BITMAP_WRITE_ERROR;
		spin_unlock_irqrestore(&bitmap->lock, flags);
366
	}
367 368 369
	if (atomic_dec_and_test(&bitmap->pending_writes))
		wake_up(&bitmap->write_wait);
}
370

371 372 373 374 375 376 377 378 379 380 381
/* copied from buffer.c */
static void
__clear_page_buffers(struct page *page)
{
	ClearPagePrivate(page);
	set_page_private(page, 0);
	page_cache_release(page);
}
static void free_buffers(struct page *page)
{
	struct buffer_head *bh = page_buffers(page);
382

383 384 385 386
	while (bh) {
		struct buffer_head *next = bh->b_this_page;
		free_buffer_head(bh);
		bh = next;
387
	}
388 389
	__clear_page_buffers(page);
	put_page(page);
390 391
}

392 393 394 395 396 397 398
/* read a page from a file.
 * We both read the page, and attach buffers to the page to record the
 * address of each block (using bmap).  These addresses will be used
 * to write the block later, completely bypassing the filesystem.
 * This usage is similar to how swap files are handled, and allows us
 * to write to a file with no concerns of memory allocation failing.
 */
399
static struct page *read_page(struct file *file, unsigned long index,
400 401
			      struct bitmap *bitmap,
			      unsigned long count)
402 403
{
	struct page *page = NULL;
J
Josef Sipek 已提交
404
	struct inode *inode = file->f_path.dentry->d_inode;
405 406
	struct buffer_head *bh;
	sector_t block;
407

408
	PRINTK("read bitmap file (%dB @ %llu)\n", (int)PAGE_SIZE,
409
			(unsigned long long)index << PAGE_SHIFT);
410

411 412 413
	page = alloc_page(GFP_KERNEL);
	if (!page)
		page = ERR_PTR(-ENOMEM);
414 415
	if (IS_ERR(page))
		goto out;
416 417 418

	bh = alloc_page_buffers(page, 1<<inode->i_blkbits, 0);
	if (!bh) {
419
		put_page(page);
420
		page = ERR_PTR(-ENOMEM);
421 422
		goto out;
	}
423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443
	attach_page_buffers(page, bh);
	block = index << (PAGE_SHIFT - inode->i_blkbits);
	while (bh) {
		if (count == 0)
			bh->b_blocknr = 0;
		else {
			bh->b_blocknr = bmap(inode, block);
			if (bh->b_blocknr == 0) {
				/* Cannot use this file! */
				free_buffers(page);
				page = ERR_PTR(-EINVAL);
				goto out;
			}
			bh->b_bdev = inode->i_sb->s_bdev;
			if (count < (1<<inode->i_blkbits))
				count = 0;
			else
				count -= (1<<inode->i_blkbits);

			bh->b_end_io = end_bitmap_write;
			bh->b_private = bitmap;
444 445 446 447
			atomic_inc(&bitmap->pending_writes);
			set_buffer_locked(bh);
			set_buffer_mapped(bh);
			submit_bh(READ, bh);
448 449 450 451 452
		}
		block++;
		bh = bh->b_this_page;
	}
	page->index = index;
453 454 455 456 457 458 459

	wait_event(bitmap->write_wait,
		   atomic_read(&bitmap->pending_writes)==0);
	if (bitmap->flags & BITMAP_WRITE_ERROR) {
		free_buffers(page);
		page = ERR_PTR(-EIO);
	}
460 461
out:
	if (IS_ERR(page))
462
		printk(KERN_ALERT "md: bitmap read error: (%dB @ %llu): %ld\n",
463 464
			(int)PAGE_SIZE,
			(unsigned long long)index << PAGE_SHIFT,
465 466 467 468 469 470 471 472 473
			PTR_ERR(page));
	return page;
}

/*
 * bitmap file superblock operations
 */

/* update the event counter and sync the superblock to disk */
474
void bitmap_update_sb(struct bitmap *bitmap)
475 476 477 478 479
{
	bitmap_super_t *sb;
	unsigned long flags;

	if (!bitmap || !bitmap->mddev) /* no bitmap for this array */
480
		return;
481 482
	if (bitmap->mddev->bitmap_info.external)
		return;
483 484 485
	spin_lock_irqsave(&bitmap->lock, flags);
	if (!bitmap->sb_page) { /* no superblock */
		spin_unlock_irqrestore(&bitmap->lock, flags);
486
		return;
487 488
	}
	spin_unlock_irqrestore(&bitmap->lock, flags);
489
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
490
	sb->events = cpu_to_le64(bitmap->mddev->events);
491 492 493 494 495
	if (bitmap->mddev->events < bitmap->events_cleared) {
		/* rocking back to read-only */
		bitmap->events_cleared = bitmap->mddev->events;
		sb->events_cleared = cpu_to_le64(bitmap->events_cleared);
	}
496 497 498
	/* Just in case these have been changed via sysfs: */
	sb->daemon_sleep = cpu_to_le32(bitmap->mddev->bitmap_info.daemon_sleep/HZ);
	sb->write_behind = cpu_to_le32(bitmap->mddev->bitmap_info.max_write_behind);
499
	kunmap_atomic(sb, KM_USER0);
500
	write_page(bitmap, bitmap->sb_page, 1);
501 502 503 504 505 506 507 508 509
}

/* print out the bitmap file superblock */
void bitmap_print_sb(struct bitmap *bitmap)
{
	bitmap_super_t *sb;

	if (!bitmap || !bitmap->sb_page)
		return;
510
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
511
	printk(KERN_DEBUG "%s: bitmap file superblock:\n", bmname(bitmap));
512 513 514
	printk(KERN_DEBUG "         magic: %08x\n", le32_to_cpu(sb->magic));
	printk(KERN_DEBUG "       version: %d\n", le32_to_cpu(sb->version));
	printk(KERN_DEBUG "          uuid: %08x.%08x.%08x.%08x\n",
515 516 517 518
					*(__u32 *)(sb->uuid+0),
					*(__u32 *)(sb->uuid+4),
					*(__u32 *)(sb->uuid+8),
					*(__u32 *)(sb->uuid+12));
519
	printk(KERN_DEBUG "        events: %llu\n",
520
			(unsigned long long) le64_to_cpu(sb->events));
521
	printk(KERN_DEBUG "events cleared: %llu\n",
522
			(unsigned long long) le64_to_cpu(sb->events_cleared));
523 524 525 526 527
	printk(KERN_DEBUG "         state: %08x\n", le32_to_cpu(sb->state));
	printk(KERN_DEBUG "     chunksize: %d B\n", le32_to_cpu(sb->chunksize));
	printk(KERN_DEBUG "  daemon sleep: %ds\n", le32_to_cpu(sb->daemon_sleep));
	printk(KERN_DEBUG "     sync size: %llu KB\n",
			(unsigned long long)le64_to_cpu(sb->sync_size)/2);
528
	printk(KERN_DEBUG "max write behind: %d\n", le32_to_cpu(sb->write_behind));
529
	kunmap_atomic(sb, KM_USER0);
530 531 532 533 534 535 536
}

/* read the superblock from the bitmap file and initialize some bitmap fields */
static int bitmap_read_sb(struct bitmap *bitmap)
{
	char *reason = NULL;
	bitmap_super_t *sb;
537
	unsigned long chunksize, daemon_sleep, write_behind;
538 539 540 541
	unsigned long long events;
	int err = -EINVAL;

	/* page 0 is the superblock, read it... */
542 543 544 545 546 547
	if (bitmap->file) {
		loff_t isize = i_size_read(bitmap->file->f_mapping->host);
		int bytes = isize > PAGE_SIZE ? PAGE_SIZE : isize;

		bitmap->sb_page = read_page(bitmap->file, 0, bitmap, bytes);
	} else {
548 549
		bitmap->sb_page = read_sb_page(bitmap->mddev,
					       bitmap->mddev->bitmap_info.offset,
550 551
					       NULL,
					       0, sizeof(bitmap_super_t));
552
	}
553 554 555 556 557 558
	if (IS_ERR(bitmap->sb_page)) {
		err = PTR_ERR(bitmap->sb_page);
		bitmap->sb_page = NULL;
		return err;
	}

559
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
560 561

	chunksize = le32_to_cpu(sb->chunksize);
562
	daemon_sleep = le32_to_cpu(sb->daemon_sleep) * HZ;
563
	write_behind = le32_to_cpu(sb->write_behind);
564 565 566 567

	/* verify that the bitmap-specific fields are valid */
	if (sb->magic != cpu_to_le32(BITMAP_MAGIC))
		reason = "bad magic";
568 569
	else if (le32_to_cpu(sb->version) < BITMAP_MAJOR_LO ||
		 le32_to_cpu(sb->version) > BITMAP_MAJOR_HI)
570
		reason = "unrecognized superblock version";
571
	else if (chunksize < 512)
572
		reason = "bitmap chunksize too small";
573 574
	else if ((1 << ffz(~chunksize)) != chunksize)
		reason = "bitmap chunksize not a power of 2";
575
	else if (daemon_sleep < 1 || daemon_sleep > MAX_SCHEDULE_TIMEOUT)
576
		reason = "daemon sleep period out of range";
577 578
	else if (write_behind > COUNTER_MAX)
		reason = "write-behind limit out of range (0 - 16383)";
579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604
	if (reason) {
		printk(KERN_INFO "%s: invalid bitmap file superblock: %s\n",
			bmname(bitmap), reason);
		goto out;
	}

	/* keep the array size field of the bitmap superblock up to date */
	sb->sync_size = cpu_to_le64(bitmap->mddev->resync_max_sectors);

	if (!bitmap->mddev->persistent)
		goto success;

	/*
	 * if we have a persistent array superblock, compare the
	 * bitmap's UUID and event counter to the mddev's
	 */
	if (memcmp(sb->uuid, bitmap->mddev->uuid, 16)) {
		printk(KERN_INFO "%s: bitmap superblock UUID mismatch\n",
			bmname(bitmap));
		goto out;
	}
	events = le64_to_cpu(sb->events);
	if (events < bitmap->mddev->events) {
		printk(KERN_INFO "%s: bitmap file is out of date (%llu < %llu) "
			"-- forcing full recovery\n", bmname(bitmap), events,
			(unsigned long long) bitmap->mddev->events);
605
		sb->state |= cpu_to_le32(BITMAP_STALE);
606 607 608
	}
success:
	/* assign fields using values from superblock */
609 610 611
	bitmap->mddev->bitmap_info.chunksize = chunksize;
	bitmap->mddev->bitmap_info.daemon_sleep = daemon_sleep;
	bitmap->mddev->bitmap_info.max_write_behind = write_behind;
612
	bitmap->flags |= le32_to_cpu(sb->state);
613 614
	if (le32_to_cpu(sb->version) == BITMAP_MAJOR_HOSTENDIAN)
		bitmap->flags |= BITMAP_HOSTENDIAN;
615
	bitmap->events_cleared = le64_to_cpu(sb->events_cleared);
616
	if (sb->state & cpu_to_le32(BITMAP_STALE))
617
		bitmap->events_cleared = bitmap->mddev->events;
618 619
	err = 0;
out:
620
	kunmap_atomic(sb, KM_USER0);
621 622 623 624 625 626 627 628 629 630
	if (err)
		bitmap_print_sb(bitmap);
	return err;
}

enum bitmap_mask_op {
	MASK_SET,
	MASK_UNSET
};

631 632 633
/* record the state of the bitmap in the superblock.  Return the old value */
static int bitmap_mask_state(struct bitmap *bitmap, enum bitmap_state bits,
			     enum bitmap_mask_op op)
634 635 636
{
	bitmap_super_t *sb;
	unsigned long flags;
637
	int old;
638 639

	spin_lock_irqsave(&bitmap->lock, flags);
640
	if (!bitmap->sb_page) { /* can't set the state */
641
		spin_unlock_irqrestore(&bitmap->lock, flags);
642
		return 0;
643 644
	}
	spin_unlock_irqrestore(&bitmap->lock, flags);
645
	sb = kmap_atomic(bitmap->sb_page, KM_USER0);
646
	old = le32_to_cpu(sb->state) & bits;
647
	switch (op) {
648 649 650 651 652 653 654 655
	case MASK_SET:
		sb->state |= cpu_to_le32(bits);
		break;
	case MASK_UNSET:
		sb->state &= cpu_to_le32(~bits);
		break;
	default:
		BUG();
656
	}
657
	kunmap_atomic(sb, KM_USER0);
658
	return old;
659 660 661 662 663 664
}

/*
 * general bitmap file operations
 */

665 666 667 668 669 670
/*
 * on-disk bitmap:
 *
 * Use one bit per "chunk" (block set). We do the disk I/O on the bitmap
 * file a page at a time. There's a superblock at the start of the file.
 */
671
/* calculate the index of the page that contains this bit */
672
static inline unsigned long file_page_index(struct bitmap *bitmap, unsigned long chunk)
673
{
674 675 676
	if (!bitmap->mddev->bitmap_info.external)
		chunk += sizeof(bitmap_super_t) << 3;
	return chunk >> PAGE_BIT_SHIFT;
677 678 679
}

/* calculate the (bit) offset of this bit within a page */
680
static inline unsigned long file_page_offset(struct bitmap *bitmap, unsigned long chunk)
681
{
682 683 684
	if (!bitmap->mddev->bitmap_info.external)
		chunk += sizeof(bitmap_super_t) << 3;
	return chunk & (PAGE_BITS - 1);
685 686 687 688 689 690 691 692 693 694 695 696
}

/*
 * return a pointer to the page in the filemap that contains the given bit
 *
 * this lookup is complicated by the fact that the bitmap sb might be exactly
 * 1 page (e.g., x86) or less than 1 page -- so the bitmap might start on page
 * 0 or page 1
 */
static inline struct page *filemap_get_page(struct bitmap *bitmap,
					unsigned long chunk)
{
697 698
	if (file_page_index(bitmap, chunk) >= bitmap->file_pages)
		return NULL;
699 700
	return bitmap->filemap[file_page_index(bitmap, chunk)
			       - file_page_index(bitmap, 0)];
701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721
}

static void bitmap_file_unmap(struct bitmap *bitmap)
{
	struct page **map, *sb_page;
	unsigned long *attr;
	int pages;
	unsigned long flags;

	spin_lock_irqsave(&bitmap->lock, flags);
	map = bitmap->filemap;
	bitmap->filemap = NULL;
	attr = bitmap->filemap_attr;
	bitmap->filemap_attr = NULL;
	pages = bitmap->file_pages;
	bitmap->file_pages = 0;
	sb_page = bitmap->sb_page;
	bitmap->sb_page = NULL;
	spin_unlock_irqrestore(&bitmap->lock, flags);

	while (pages--)
722
		if (map[pages] != sb_page) /* 0 is sb_page, release it below */
723
			free_buffers(map[pages]);
724 725 726
	kfree(map);
	kfree(attr);

727 728
	if (sb_page)
		free_buffers(sb_page);
729 730 731 732 733 734 735 736 737 738 739 740
}

static void bitmap_file_put(struct bitmap *bitmap)
{
	struct file *file;
	unsigned long flags;

	spin_lock_irqsave(&bitmap->lock, flags);
	file = bitmap->file;
	bitmap->file = NULL;
	spin_unlock_irqrestore(&bitmap->lock, flags);

741 742 743
	if (file)
		wait_event(bitmap->write_wait,
			   atomic_read(&bitmap->pending_writes)==0);
744 745
	bitmap_file_unmap(bitmap);

746
	if (file) {
J
Josef Sipek 已提交
747
		struct inode *inode = file->f_path.dentry->d_inode;
748
		invalidate_mapping_pages(inode->i_mapping, 0, -1);
749
		fput(file);
750
	}
751 752 753 754 755 756 757 758 759 760 761
}

/*
 * bitmap_file_kick - if an error occurs while manipulating the bitmap file
 * then it is no longer reliable, so we stop using it and we mark the file
 * as failed in the superblock
 */
static void bitmap_file_kick(struct bitmap *bitmap)
{
	char *path, *ptr = NULL;

762 763
	if (bitmap_mask_state(bitmap, BITMAP_STALE, MASK_SET) == 0) {
		bitmap_update_sb(bitmap);
764

765 766 767
		if (bitmap->file) {
			path = kmalloc(PAGE_SIZE, GFP_KERNEL);
			if (path)
C
Christoph Hellwig 已提交
768 769 770
				ptr = d_path(&bitmap->file->f_path, path,
					     PAGE_SIZE);

771 772
			printk(KERN_ALERT
			      "%s: kicking failed bitmap file %s from array!\n",
C
Christoph Hellwig 已提交
773
			      bmname(bitmap), IS_ERR(ptr) ? "" : ptr);
774

775 776 777 778 779
			kfree(path);
		} else
			printk(KERN_ALERT
			       "%s: disabling internal bitmap due to errors\n",
			       bmname(bitmap));
780
	}
781 782 783 784 785 786 787

	bitmap_file_put(bitmap);

	return;
}

enum bitmap_page_attr {
788 789 790
	BITMAP_PAGE_DIRTY = 0,     /* there are set bits that need to be synced */
	BITMAP_PAGE_CLEAN = 1,     /* there are bits that might need to be cleared */
	BITMAP_PAGE_NEEDWRITE = 2, /* there are cleared bits that need to be synced */
791 792 793 794 795
};

static inline void set_page_attr(struct bitmap *bitmap, struct page *page,
				enum bitmap_page_attr attr)
{
796
	__set_bit((page->index<<2) + attr, bitmap->filemap_attr);
797 798 799 800 801
}

static inline void clear_page_attr(struct bitmap *bitmap, struct page *page,
				enum bitmap_page_attr attr)
{
802
	__clear_bit((page->index<<2) + attr, bitmap->filemap_attr);
803 804
}

805 806
static inline unsigned long test_page_attr(struct bitmap *bitmap, struct page *page,
					   enum bitmap_page_attr attr)
807
{
808
	return test_bit((page->index<<2) + attr, bitmap->filemap_attr);
809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824
}

/*
 * bitmap_file_set_bit -- called before performing a write to the md device
 * to set (and eventually sync) a particular bit in the bitmap file
 *
 * we set the bit immediately, then we record the page number so that
 * when an unplug occurs, we can flush the dirty pages out to disk
 */
static void bitmap_file_set_bit(struct bitmap *bitmap, sector_t block)
{
	unsigned long bit;
	struct page *page;
	void *kaddr;
	unsigned long chunk = block >> CHUNK_BLOCK_SHIFT(bitmap);

825
	if (!bitmap->filemap)
826 827 828
		return;

	page = filemap_get_page(bitmap, chunk);
829 830
	if (!page)
		return;
831
	bit = file_page_offset(bitmap, chunk);
832

833
	/* set the bit */
834
	kaddr = kmap_atomic(page, KM_USER0);
835 836 837 838
	if (bitmap->flags & BITMAP_HOSTENDIAN)
		set_bit(bit, kaddr);
	else
		ext2_set_bit(bit, kaddr);
839 840 841 842 843 844 845 846 847 848
	kunmap_atomic(kaddr, KM_USER0);
	PRINTK("set file bit %lu page %lu\n", bit, page->index);

	/* record page number so it gets flushed to disk when unplug occurs */
	set_page_attr(bitmap, page, BITMAP_PAGE_DIRTY);
}

/* this gets called when the md device is ready to unplug its underlying
 * (slave) device queues -- before we let any writes go down, we need to
 * sync the dirty pages of the bitmap file to disk */
849
void bitmap_unplug(struct bitmap *bitmap)
850
{
851 852
	unsigned long i, flags;
	int dirty, need_write;
853 854 855 856
	struct page *page;
	int wait = 0;

	if (!bitmap)
857
		return;
858 859 860 861 862

	/* look at each page to see if there are any set bits that need to be
	 * flushed out to disk */
	for (i = 0; i < bitmap->file_pages; i++) {
		spin_lock_irqsave(&bitmap->lock, flags);
863
		if (!bitmap->filemap) {
864
			spin_unlock_irqrestore(&bitmap->lock, flags);
865
			return;
866 867
		}
		page = bitmap->filemap[i];
868 869
		dirty = test_page_attr(bitmap, page, BITMAP_PAGE_DIRTY);
		need_write = test_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE);
870 871
		clear_page_attr(bitmap, page, BITMAP_PAGE_DIRTY);
		clear_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE);
872
		if (dirty)
873 874 875
			wait = 1;
		spin_unlock_irqrestore(&bitmap->lock, flags);

876
		if (dirty || need_write)
877
			write_page(bitmap, page, 0);
878 879
	}
	if (wait) { /* if any writes were performed, we need to wait on them */
880
		if (bitmap->file)
881 882
			wait_event(bitmap->write_wait,
				   atomic_read(&bitmap->pending_writes)==0);
883
		else
884
			md_super_wait(bitmap->mddev);
885
	}
886 887
	if (bitmap->flags & BITMAP_WRITE_ERROR)
		bitmap_file_kick(bitmap);
888
}
889
EXPORT_SYMBOL(bitmap_unplug);
890

891
static void bitmap_set_memory_bits(struct bitmap *bitmap, sector_t offset, int needed);
892 893 894 895 896 897 898
/* * bitmap_init_from_disk -- called at bitmap_create time to initialize
 * the in-memory bitmap from the on-disk bitmap -- also, sets up the
 * memory mapping of the bitmap file
 * Special cases:
 *   if there's no bitmap file, or if the bitmap file had been
 *   previously kicked from the array, we mark all the bits as
 *   1's in order to cause a full resync.
899 900 901
 *
 * We ignore all bits for sectors that end earlier than 'start'.
 * This is used when reading an out-of-date bitmap...
902
 */
903
static int bitmap_init_from_disk(struct bitmap *bitmap, sector_t start)
904 905 906 907 908
{
	unsigned long i, chunks, index, oldindex, bit;
	struct page *page = NULL, *oldpage = NULL;
	unsigned long num_pages, bit_cnt = 0;
	struct file *file;
909
	unsigned long bytes, offset;
910 911
	int outofdate;
	int ret = -ENOSPC;
912
	void *paddr;
913 914 915 916

	chunks = bitmap->chunks;
	file = bitmap->file;

917
	BUG_ON(!file && !bitmap->mddev->bitmap_info.offset);
918

919
#ifdef INJECT_FAULTS_3
920 921 922 923 924 925 926 927 928
	outofdate = 1;
#else
	outofdate = bitmap->flags & BITMAP_STALE;
#endif
	if (outofdate)
		printk(KERN_INFO "%s: bitmap file is out of date, doing full "
			"recovery\n", bmname(bitmap));

	bytes = (chunks + 7) / 8;
929 930
	if (!bitmap->mddev->bitmap_info.external)
		bytes += sizeof(bitmap_super_t);
931

932
	num_pages = (bytes + PAGE_SIZE - 1) / PAGE_SIZE;
933

934
	if (file && i_size_read(file->f_mapping->host) < bytes) {
935 936 937
		printk(KERN_INFO "%s: bitmap file too short %lu < %lu\n",
			bmname(bitmap),
			(unsigned long) i_size_read(file->f_mapping->host),
938
			bytes);
939
		goto err;
940
	}
941 942 943

	ret = -ENOMEM;

944
	bitmap->filemap = kmalloc(sizeof(struct page *) * num_pages, GFP_KERNEL);
945
	if (!bitmap->filemap)
946
		goto err;
947

948 949
	/* We need 4 bits per page, rounded up to a multiple of sizeof(unsigned long) */
	bitmap->filemap_attr = kzalloc(
950
		roundup(DIV_ROUND_UP(num_pages*4, 8), sizeof(unsigned long)),
951
		GFP_KERNEL);
952
	if (!bitmap->filemap_attr)
953
		goto err;
954 955 956 957

	oldindex = ~0L;

	for (i = 0; i < chunks; i++) {
958
		int b;
959 960
		index = file_page_index(bitmap, i);
		bit = file_page_offset(bitmap, i);
961
		if (index != oldindex) { /* this is a new page, read it in */
962
			int count;
963
			/* unmap the old page, we're done with it */
964
			if (index == num_pages-1)
965
				count = bytes - index * PAGE_SIZE;
966 967
			else
				count = PAGE_SIZE;
968
			if (index == 0 && bitmap->sb_page) {
969 970 971 972 973 974 975
				/*
				 * if we're here then the superblock page
				 * contains some bits (PAGE_SIZE != sizeof sb)
				 * we've already read it in, so just use it
				 */
				page = bitmap->sb_page;
				offset = sizeof(bitmap_super_t);
N
NeilBrown 已提交
976 977
				if (!file)
					read_sb_page(bitmap->mddev,
978
						     bitmap->mddev->bitmap_info.offset,
N
NeilBrown 已提交
979 980
						     page,
						     index, count);
981
			} else if (file) {
982
				page = read_page(file, index, bitmap, count);
983 984
				offset = 0;
			} else {
985 986
				page = read_sb_page(bitmap->mddev,
						    bitmap->mddev->bitmap_info.offset,
987 988
						    NULL,
						    index, count);
989 990
				offset = 0;
			}
991 992
			if (IS_ERR(page)) { /* read error */
				ret = PTR_ERR(page);
993
				goto err;
994 995
			}

996 997 998
			oldindex = index;
			oldpage = page;

999 1000 1001
			bitmap->filemap[bitmap->file_pages++] = page;
			bitmap->last_page_size = count;

1002 1003 1004
			if (outofdate) {
				/*
				 * if bitmap is out of date, dirty the
1005
				 * whole page and write it out
1006
				 */
1007 1008
				paddr = kmap_atomic(page, KM_USER0);
				memset(paddr + offset, 0xff,
1009
				       PAGE_SIZE - offset);
1010
				kunmap_atomic(paddr, KM_USER0);
1011 1012 1013
				write_page(bitmap, page, 1);

				ret = -EIO;
1014
				if (bitmap->flags & BITMAP_WRITE_ERROR)
1015
					goto err;
1016 1017
			}
		}
1018
		paddr = kmap_atomic(page, KM_USER0);
1019
		if (bitmap->flags & BITMAP_HOSTENDIAN)
1020
			b = test_bit(bit, paddr);
1021
		else
1022 1023
			b = ext2_test_bit(bit, paddr);
		kunmap_atomic(paddr, KM_USER0);
1024
		if (b) {
1025
			/* if the disk bit is set, set the memory bit */
1026 1027 1028 1029 1030
			int needed = ((sector_t)(i+1) << (CHUNK_BLOCK_SHIFT(bitmap))
				      >= start);
			bitmap_set_memory_bits(bitmap,
					       (sector_t)i << CHUNK_BLOCK_SHIFT(bitmap),
					       needed);
1031
			bit_cnt++;
1032
			set_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
1033 1034 1035
		}
	}

1036
	/* everything went OK */
1037 1038 1039 1040 1041 1042 1043 1044 1045
	ret = 0;
	bitmap_mask_state(bitmap, BITMAP_STALE, MASK_UNSET);

	if (bit_cnt) { /* Kick recovery if any bits were set */
		set_bit(MD_RECOVERY_NEEDED, &bitmap->mddev->recovery);
		md_wakeup_thread(bitmap->mddev->thread);
	}

	printk(KERN_INFO "%s: bitmap initialized from disk: "
1046 1047 1048 1049
		"read %lu/%lu pages, set %lu bits\n",
		bmname(bitmap), bitmap->file_pages, num_pages, bit_cnt);

	return 0;
1050

1051 1052 1053
 err:
	printk(KERN_INFO "%s: bitmap initialisation failed: %d\n",
	       bmname(bitmap), ret);
1054 1055 1056
	return ret;
}

1057 1058 1059 1060 1061
void bitmap_write_all(struct bitmap *bitmap)
{
	/* We don't actually write all bitmap blocks here,
	 * just flag them as needing to be written
	 */
1062
	int i;
1063

1064
	for (i = 0; i < bitmap->file_pages; i++)
1065 1066
		set_page_attr(bitmap, bitmap->filemap[i],
			      BITMAP_PAGE_NEEDWRITE);
1067 1068
}

1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084
static void bitmap_count_page(struct bitmap *bitmap, sector_t offset, int inc)
{
	sector_t chunk = offset >> CHUNK_BLOCK_SHIFT(bitmap);
	unsigned long page = chunk >> PAGE_COUNTER_SHIFT;
	bitmap->bp[page].count += inc;
	bitmap_checkfree(bitmap, page);
}
static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
					    sector_t offset, int *blocks,
					    int create);

/*
 * bitmap daemon -- periodically wakes up to clean bits and flush pages
 *			out to disk
 */

1085
void bitmap_daemon_work(mddev_t *mddev)
1086
{
1087
	struct bitmap *bitmap;
1088
	unsigned long j;
1089 1090 1091
	unsigned long flags;
	struct page *page = NULL, *lastpage = NULL;
	int blocks;
1092
	void *paddr;
1093

1094 1095 1096
	/* Use a mutex to guard daemon_work against
	 * bitmap_destroy.
	 */
1097
	mutex_lock(&mddev->bitmap_info.mutex);
1098 1099
	bitmap = mddev->bitmap;
	if (bitmap == NULL) {
1100
		mutex_unlock(&mddev->bitmap_info.mutex);
1101
		return;
1102
	}
1103
	if (time_before(jiffies, bitmap->daemon_lastrun
1104
			+ bitmap->mddev->bitmap_info.daemon_sleep))
1105 1106
		goto done;

1107
	bitmap->daemon_lastrun = jiffies;
1108 1109
	if (bitmap->allclean) {
		bitmap->mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;
1110
		goto done;
1111 1112
	}
	bitmap->allclean = 1;
1113

1114
	spin_lock_irqsave(&bitmap->lock, flags);
1115 1116
	for (j = 0; j < bitmap->chunks; j++) {
		bitmap_counter_t *bmc;
1117
		if (!bitmap->filemap)
1118 1119 1120 1121 1122 1123
			/* error or shutdown */
			break;

		page = filemap_get_page(bitmap, j);

		if (page != lastpage) {
1124
			/* skip this page unless it's marked as needing cleaning */
1125 1126 1127
			if (!test_page_attr(bitmap, page, BITMAP_PAGE_CLEAN)) {
				int need_write = test_page_attr(bitmap, page,
								BITMAP_PAGE_NEEDWRITE);
1128
				if (need_write)
1129
					clear_page_attr(bitmap, page, BITMAP_PAGE_NEEDWRITE);
1130

1131
				spin_unlock_irqrestore(&bitmap->lock, flags);
1132
				if (need_write) {
1133
					write_page(bitmap, page, 0);
1134 1135
					bitmap->allclean = 0;
				}
1136 1137
				spin_lock_irqsave(&bitmap->lock, flags);
				j |= (PAGE_BITS - 1);
1138 1139 1140
				continue;
			}

1141 1142
			/* grab the new page, sync and release the old */
			if (lastpage != NULL) {
1143
				if (test_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE)) {
1144 1145
					clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
					spin_unlock_irqrestore(&bitmap->lock, flags);
1146
					write_page(bitmap, lastpage, 0);
1147 1148 1149 1150 1151 1152 1153
				} else {
					set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
					spin_unlock_irqrestore(&bitmap->lock, flags);
				}
			} else
				spin_unlock_irqrestore(&bitmap->lock, flags);
			lastpage = page;
1154 1155 1156 1157

			/* We are possibly going to clear some bits, so make
			 * sure that events_cleared is up-to-date.
			 */
1158 1159
			if (bitmap->need_sync &&
			    bitmap->mddev->bitmap_info.external == 0) {
1160 1161 1162 1163 1164 1165 1166 1167
				bitmap_super_t *sb;
				bitmap->need_sync = 0;
				sb = kmap_atomic(bitmap->sb_page, KM_USER0);
				sb->events_cleared =
					cpu_to_le64(bitmap->events_cleared);
				kunmap_atomic(sb, KM_USER0);
				write_page(bitmap, bitmap->sb_page, 1);
			}
1168
			spin_lock_irqsave(&bitmap->lock, flags);
1169 1170
			if (!bitmap->need_sync)
				clear_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
1171
		}
1172 1173 1174
		bmc = bitmap_get_counter(bitmap,
					 (sector_t)j << CHUNK_BLOCK_SHIFT(bitmap),
					 &blocks, 0);
1175
		if (bmc) {
1176 1177 1178
			if (*bmc)
				bitmap->allclean = 0;

1179
			if (*bmc == 2) {
1180
				*bmc = 1; /* maybe clear the bit next time */
1181
				set_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
1182
			} else if (*bmc == 1 && !bitmap->need_sync) {
1183 1184
				/* we can clear the bit */
				*bmc = 0;
1185 1186
				bitmap_count_page(bitmap,
						  (sector_t)j << CHUNK_BLOCK_SHIFT(bitmap),
1187 1188 1189
						  -1);

				/* clear the bit */
1190
				paddr = kmap_atomic(page, KM_USER0);
1191
				if (bitmap->flags & BITMAP_HOSTENDIAN)
1192 1193
					clear_bit(file_page_offset(bitmap, j),
						  paddr);
1194
				else
1195 1196
					ext2_clear_bit(file_page_offset(bitmap, j),
						       paddr);
1197
				kunmap_atomic(paddr, KM_USER0);
1198
			}
1199 1200
		} else
			j |= PAGE_COUNTER_MASK;
1201
	}
1202
	spin_unlock_irqrestore(&bitmap->lock, flags);
1203 1204 1205 1206

	/* now sync the final page */
	if (lastpage != NULL) {
		spin_lock_irqsave(&bitmap->lock, flags);
1207
		if (test_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE)) {
1208 1209
			clear_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
			spin_unlock_irqrestore(&bitmap->lock, flags);
1210
			write_page(bitmap, lastpage, 0);
1211 1212 1213 1214 1215 1216
		} else {
			set_page_attr(bitmap, lastpage, BITMAP_PAGE_NEEDWRITE);
			spin_unlock_irqrestore(&bitmap->lock, flags);
		}
	}

1217
 done:
1218
	if (bitmap->allclean == 0)
1219
		bitmap->mddev->thread->timeout =
1220
			bitmap->mddev->bitmap_info.daemon_sleep;
1221
	mutex_unlock(&mddev->bitmap_info.mutex);
1222 1223 1224 1225 1226
}

static bitmap_counter_t *bitmap_get_counter(struct bitmap *bitmap,
					    sector_t offset, int *blocks,
					    int create)
1227 1228
__releases(bitmap->lock)
__acquires(bitmap->lock)
1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240
{
	/* If 'create', we might release the lock and reclaim it.
	 * The lock must have been taken with interrupts enabled.
	 * If !create, we don't release the lock.
	 */
	sector_t chunk = offset >> CHUNK_BLOCK_SHIFT(bitmap);
	unsigned long page = chunk >> PAGE_COUNTER_SHIFT;
	unsigned long pageoff = (chunk & PAGE_COUNTER_MASK) << COUNTER_BYTE_SHIFT;
	sector_t csize;

	if (bitmap_checkpage(bitmap, page, create) < 0) {
		csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap));
1241
		*blocks = csize - (offset & (csize - 1));
1242 1243 1244 1245 1246 1247 1248 1249 1250 1251
		return NULL;
	}
	/* now locked ... */

	if (bitmap->bp[page].hijacked) { /* hijacked pointer */
		/* should we use the first or second counter field
		 * of the hijacked pointer? */
		int hi = (pageoff > PAGE_COUNTER_MASK);
		csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap) +
					  PAGE_COUNTER_SHIFT - 1);
1252
		*blocks = csize - (offset & (csize - 1));
1253 1254 1255 1256
		return  &((bitmap_counter_t *)
			  &bitmap->bp[page].map)[hi];
	} else { /* page is allocated */
		csize = ((sector_t)1) << (CHUNK_BLOCK_SHIFT(bitmap));
1257
		*blocks = csize - (offset & (csize - 1));
1258 1259 1260 1261 1262
		return (bitmap_counter_t *)
			&(bitmap->bp[page].map[pageoff]);
	}
}

1263
int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors, int behind)
1264
{
1265 1266
	if (!bitmap)
		return 0;
1267 1268

	if (behind) {
1269
		int bw;
1270
		atomic_inc(&bitmap->behind_writes);
1271 1272 1273 1274
		bw = atomic_read(&bitmap->behind_writes);
		if (bw > bitmap->behind_writes_used)
			bitmap->behind_writes_used = bw;

1275
		PRINTK(KERN_DEBUG "inc write-behind count %d/%d\n",
1276
		       bw, bitmap->max_write_behind);
1277 1278
	}

1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289
	while (sectors) {
		int blocks;
		bitmap_counter_t *bmc;

		spin_lock_irq(&bitmap->lock);
		bmc = bitmap_get_counter(bitmap, offset, &blocks, 1);
		if (!bmc) {
			spin_unlock_irq(&bitmap->lock);
			return 0;
		}

1290 1291 1292 1293 1294 1295 1296 1297 1298
		if (unlikely((*bmc & COUNTER_MAX) == COUNTER_MAX)) {
			DEFINE_WAIT(__wait);
			/* note that it is safe to do the prepare_to_wait
			 * after the test as long as we do it before dropping
			 * the spinlock.
			 */
			prepare_to_wait(&bitmap->overflow_wait, &__wait,
					TASK_UNINTERRUPTIBLE);
			spin_unlock_irq(&bitmap->lock);
1299
			blk_unplug(bitmap->mddev->queue);
1300 1301 1302 1303 1304
			schedule();
			finish_wait(&bitmap->overflow_wait, &__wait);
			continue;
		}

1305
		switch (*bmc) {
1306 1307
		case 0:
			bitmap_file_set_bit(bitmap, offset);
1308
			bitmap_count_page(bitmap, offset, 1);
1309
			blk_plug_device_unlocked(bitmap->mddev->queue);
1310 1311 1312 1313
			/* fall through */
		case 1:
			*bmc = 2;
		}
1314

1315 1316 1317 1318 1319 1320 1321
		(*bmc)++;

		spin_unlock_irq(&bitmap->lock);

		offset += blocks;
		if (sectors > blocks)
			sectors -= blocks;
1322 1323
		else
			sectors = 0;
1324
	}
1325
	bitmap->allclean = 0;
1326 1327
	return 0;
}
1328
EXPORT_SYMBOL(bitmap_startwrite);
1329 1330

void bitmap_endwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors,
1331
		     int success, int behind)
1332
{
1333 1334
	if (!bitmap)
		return;
1335
	if (behind) {
1336 1337
		if (atomic_dec_and_test(&bitmap->behind_writes))
			wake_up(&bitmap->behind_wait);
1338 1339 1340
		PRINTK(KERN_DEBUG "dec write-behind count %d/%d\n",
		  atomic_read(&bitmap->behind_writes), bitmap->max_write_behind);
	}
1341 1342 1343
	if (bitmap->mddev->degraded)
		/* Never clear bits or update events_cleared when degraded */
		success = 0;
1344

1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356
	while (sectors) {
		int blocks;
		unsigned long flags;
		bitmap_counter_t *bmc;

		spin_lock_irqsave(&bitmap->lock, flags);
		bmc = bitmap_get_counter(bitmap, offset, &blocks, 0);
		if (!bmc) {
			spin_unlock_irqrestore(&bitmap->lock, flags);
			return;
		}

1357 1358 1359 1360
		if (success &&
		    bitmap->events_cleared < bitmap->mddev->events) {
			bitmap->events_cleared = bitmap->mddev->events;
			bitmap->need_sync = 1;
1361
			sysfs_notify_dirent(bitmap->sysfs_can_clear);
1362 1363
		}

1364 1365 1366
		if (!success && ! (*bmc & NEEDED_MASK))
			*bmc |= NEEDED_MASK;

1367 1368 1369
		if ((*bmc & COUNTER_MAX) == COUNTER_MAX)
			wake_up(&bitmap->overflow_wait);

1370
		(*bmc)--;
1371
		if (*bmc <= 2)
1372 1373 1374
			set_page_attr(bitmap,
				      filemap_get_page(bitmap, offset >> CHUNK_BLOCK_SHIFT(bitmap)),
				      BITMAP_PAGE_CLEAN);
1375

1376 1377 1378 1379
		spin_unlock_irqrestore(&bitmap->lock, flags);
		offset += blocks;
		if (sectors > blocks)
			sectors -= blocks;
1380 1381
		else
			sectors = 0;
1382 1383
	}
}
1384
EXPORT_SYMBOL(bitmap_endwrite);
1385

1386 1387
static int __bitmap_start_sync(struct bitmap *bitmap, sector_t offset, int *blocks,
			       int degraded)
1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403
{
	bitmap_counter_t *bmc;
	int rv;
	if (bitmap == NULL) {/* FIXME or bitmap set as 'failed' */
		*blocks = 1024;
		return 1; /* always resync if no bitmap */
	}
	spin_lock_irq(&bitmap->lock);
	bmc = bitmap_get_counter(bitmap, offset, blocks, 0);
	rv = 0;
	if (bmc) {
		/* locked */
		if (RESYNC(*bmc))
			rv = 1;
		else if (NEEDED(*bmc)) {
			rv = 1;
1404 1405 1406 1407
			if (!degraded) { /* don't set/clear bits if degraded */
				*bmc |= RESYNC_MASK;
				*bmc &= ~NEEDED_MASK;
			}
1408 1409 1410
		}
	}
	spin_unlock_irq(&bitmap->lock);
1411
	bitmap->allclean = 0;
1412 1413 1414
	return rv;
}

1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436
int bitmap_start_sync(struct bitmap *bitmap, sector_t offset, int *blocks,
		      int degraded)
{
	/* bitmap_start_sync must always report on multiples of whole
	 * pages, otherwise resync (which is very PAGE_SIZE based) will
	 * get confused.
	 * So call __bitmap_start_sync repeatedly (if needed) until
	 * At least PAGE_SIZE>>9 blocks are covered.
	 * Return the 'or' of the result.
	 */
	int rv = 0;
	int blocks1;

	*blocks = 0;
	while (*blocks < (PAGE_SIZE>>9)) {
		rv |= __bitmap_start_sync(bitmap, offset,
					  &blocks1, degraded);
		offset += blocks1;
		*blocks += blocks1;
	}
	return rv;
}
1437
EXPORT_SYMBOL(bitmap_start_sync);
1438

1439 1440 1441 1442
void bitmap_end_sync(struct bitmap *bitmap, sector_t offset, int *blocks, int aborted)
{
	bitmap_counter_t *bmc;
	unsigned long flags;
1443 1444

	if (bitmap == NULL) {
1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458
		*blocks = 1024;
		return;
	}
	spin_lock_irqsave(&bitmap->lock, flags);
	bmc = bitmap_get_counter(bitmap, offset, blocks, 0);
	if (bmc == NULL)
		goto unlock;
	/* locked */
	if (RESYNC(*bmc)) {
		*bmc &= ~RESYNC_MASK;

		if (!NEEDED(*bmc) && aborted)
			*bmc |= NEEDED_MASK;
		else {
1459
			if (*bmc <= 2)
1460 1461 1462 1463 1464 1465 1466
				set_page_attr(bitmap,
					      filemap_get_page(bitmap, offset >> CHUNK_BLOCK_SHIFT(bitmap)),
					      BITMAP_PAGE_CLEAN);
		}
	}
 unlock:
	spin_unlock_irqrestore(&bitmap->lock, flags);
1467
	bitmap->allclean = 0;
1468
}
1469
EXPORT_SYMBOL(bitmap_end_sync);
1470 1471 1472 1473 1474 1475 1476 1477 1478

void bitmap_close_sync(struct bitmap *bitmap)
{
	/* Sync has finished, and any bitmap chunks that weren't synced
	 * properly have been aborted.  It remains to us to clear the
	 * RESYNC bit wherever it is still on
	 */
	sector_t sector = 0;
	int blocks;
N
NeilBrown 已提交
1479 1480
	if (!bitmap)
		return;
1481 1482
	while (sector < bitmap->mddev->resync_max_sectors) {
		bitmap_end_sync(bitmap, sector, &blocks, 0);
N
NeilBrown 已提交
1483 1484 1485
		sector += blocks;
	}
}
1486
EXPORT_SYMBOL(bitmap_close_sync);
N
NeilBrown 已提交
1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499

void bitmap_cond_end_sync(struct bitmap *bitmap, sector_t sector)
{
	sector_t s = 0;
	int blocks;

	if (!bitmap)
		return;
	if (sector == 0) {
		bitmap->last_end_sync = jiffies;
		return;
	}
	if (time_before(jiffies, (bitmap->last_end_sync
1500
				  + bitmap->mddev->bitmap_info.daemon_sleep)))
N
NeilBrown 已提交
1501 1502 1503 1504
		return;
	wait_event(bitmap->mddev->recovery_wait,
		   atomic_read(&bitmap->mddev->recovery_active) == 0);

1505
	bitmap->mddev->curr_resync_completed = bitmap->mddev->curr_resync;
1506 1507
	if (bitmap->mddev->persistent)
		set_bit(MD_CHANGE_CLEAN, &bitmap->mddev->flags);
N
NeilBrown 已提交
1508 1509 1510 1511 1512
	sector &= ~((1ULL << CHUNK_BLOCK_SHIFT(bitmap)) - 1);
	s = 0;
	while (s < sector && s < bitmap->mddev->resync_max_sectors) {
		bitmap_end_sync(bitmap, s, &blocks, 0);
		s += blocks;
1513
	}
N
NeilBrown 已提交
1514
	bitmap->last_end_sync = jiffies;
1515
	sysfs_notify(&bitmap->mddev->kobj, NULL, "sync_completed");
1516
}
1517
EXPORT_SYMBOL(bitmap_cond_end_sync);
1518

1519
static void bitmap_set_memory_bits(struct bitmap *bitmap, sector_t offset, int needed)
1520 1521
{
	/* For each chunk covered by any of these sectors, set the
1522
	 * counter to 1 and set resync_needed.  They should all
1523 1524
	 * be 0 at this point
	 */
1525 1526 1527 1528 1529 1530

	int secs;
	bitmap_counter_t *bmc;
	spin_lock_irq(&bitmap->lock);
	bmc = bitmap_get_counter(bitmap, offset, &secs, 1);
	if (!bmc) {
1531
		spin_unlock_irq(&bitmap->lock);
1532
		return;
1533
	}
1534
	if (!*bmc) {
1535
		struct page *page;
1536
		*bmc = 1 | (needed ? NEEDED_MASK : 0);
1537 1538 1539 1540 1541
		bitmap_count_page(bitmap, offset, 1);
		page = filemap_get_page(bitmap, offset >> CHUNK_BLOCK_SHIFT(bitmap));
		set_page_attr(bitmap, page, BITMAP_PAGE_CLEAN);
	}
	spin_unlock_irq(&bitmap->lock);
1542
	bitmap->allclean = 0;
1543 1544
}

1545 1546 1547 1548 1549 1550
/* dirty the memory and file bits for bitmap chunks "s" to "e" */
void bitmap_dirty_bits(struct bitmap *bitmap, unsigned long s, unsigned long e)
{
	unsigned long chunk;

	for (chunk = s; chunk <= e; chunk++) {
1551
		sector_t sec = (sector_t)chunk << CHUNK_BLOCK_SHIFT(bitmap);
1552 1553
		bitmap_set_memory_bits(bitmap, sec, 1);
		bitmap_file_set_bit(bitmap, sec);
1554 1555 1556 1557 1558 1559
		if (sec < bitmap->mddev->recovery_cp)
			/* We are asserting that the array is dirty,
			 * so move the recovery_cp address back so
			 * that it is obvious that it is dirty
			 */
			bitmap->mddev->recovery_cp = sec;
1560 1561 1562
	}
}

1563 1564 1565 1566 1567 1568
/*
 * flush out any pending updates
 */
void bitmap_flush(mddev_t *mddev)
{
	struct bitmap *bitmap = mddev->bitmap;
1569
	long sleep;
1570 1571 1572 1573 1574 1575 1576

	if (!bitmap) /* there was no bitmap */
		return;

	/* run the daemon_work three time to ensure everything is flushed
	 * that can be
	 */
1577
	sleep = mddev->bitmap_info.daemon_sleep * 2;
1578
	bitmap->daemon_lastrun -= sleep;
1579
	bitmap_daemon_work(mddev);
1580
	bitmap->daemon_lastrun -= sleep;
1581
	bitmap_daemon_work(mddev);
1582
	bitmap->daemon_lastrun -= sleep;
1583
	bitmap_daemon_work(mddev);
1584 1585 1586
	bitmap_update_sb(bitmap);
}

1587 1588 1589
/*
 * free memory that was allocated
 */
1590
static void bitmap_free(struct bitmap *bitmap)
1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612
{
	unsigned long k, pages;
	struct bitmap_page *bp;

	if (!bitmap) /* there was no bitmap */
		return;

	/* release the bitmap file and kill the daemon */
	bitmap_file_put(bitmap);

	bp = bitmap->bp;
	pages = bitmap->pages;

	/* free all allocated memory */

	if (bp) /* deallocate the page memory */
		for (k = 0; k < pages; k++)
			if (bp[k].map && !bp[k].hijacked)
				kfree(bp[k].map);
	kfree(bp);
	kfree(bitmap);
}
1613

1614 1615 1616 1617 1618 1619 1620
void bitmap_destroy(mddev_t *mddev)
{
	struct bitmap *bitmap = mddev->bitmap;

	if (!bitmap) /* there was no bitmap */
		return;

1621
	mutex_lock(&mddev->bitmap_info.mutex);
1622
	mddev->bitmap = NULL; /* disconnect from the md device */
1623
	mutex_unlock(&mddev->bitmap_info.mutex);
1624 1625
	if (mddev->thread)
		mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;
1626

1627 1628 1629
	if (bitmap->sysfs_can_clear)
		sysfs_put(bitmap->sysfs_can_clear);

1630 1631
	bitmap_free(bitmap);
}
1632 1633 1634 1635 1636 1637 1638 1639

/*
 * initialize the bitmap structure
 * if this returns an error, bitmap_destroy must be called to do clean up
 */
int bitmap_create(mddev_t *mddev)
{
	struct bitmap *bitmap;
1640
	sector_t blocks = mddev->resync_max_sectors;
1641 1642
	unsigned long chunks;
	unsigned long pages;
1643
	struct file *file = mddev->bitmap_info.file;
1644
	int err;
1645
	sector_t start;
1646
	struct sysfs_dirent *bm;
1647

A
Alexey Dobriyan 已提交
1648
	BUILD_BUG_ON(sizeof(bitmap_super_t) != 256);
1649

1650
	if (!file && !mddev->bitmap_info.offset) /* bitmap disabled, nothing to do */
1651 1652
		return 0;

1653
	BUG_ON(file && mddev->bitmap_info.offset);
1654

1655
	bitmap = kzalloc(sizeof(*bitmap), GFP_KERNEL);
1656 1657 1658 1659
	if (!bitmap)
		return -ENOMEM;

	spin_lock_init(&bitmap->lock);
1660 1661
	atomic_set(&bitmap->pending_writes, 0);
	init_waitqueue_head(&bitmap->write_wait);
1662
	init_waitqueue_head(&bitmap->overflow_wait);
1663
	init_waitqueue_head(&bitmap->behind_wait);
1664

1665 1666
	bitmap->mddev = mddev;

1667
	bm = sysfs_get_dirent(mddev->kobj.sd, NULL, "bitmap");
1668
	if (bm) {
1669
		bitmap->sysfs_can_clear = sysfs_get_dirent(bm, NULL, "can_clear");
1670 1671 1672 1673
		sysfs_put(bm);
	} else
		bitmap->sysfs_can_clear = NULL;

1674
	bitmap->file = file;
1675 1676
	if (file) {
		get_file(file);
1677 1678 1679 1680
		/* As future accesses to this file will use bmap,
		 * and bypass the page cache, we must sync the file
		 * first.
		 */
1681
		vfs_fsync(file, 1);
1682
	}
1683
	/* read superblock from bitmap file (this sets mddev->bitmap_info.chunksize) */
1684 1685 1686 1687 1688 1689 1690 1691 1692 1693
	if (!mddev->bitmap_info.external)
		err = bitmap_read_sb(bitmap);
	else {
		err = 0;
		if (mddev->bitmap_info.chunksize == 0 ||
		    mddev->bitmap_info.daemon_sleep == 0)
			/* chunksize and time_base need to be
			 * set first. */
			err = -EINVAL;
	}
1694
	if (err)
1695
		goto error;
1696

1697
	bitmap->daemon_lastrun = jiffies;
1698
	bitmap->chunkshift = ffz(~mddev->bitmap_info.chunksize);
1699 1700

	/* now that chunksize and chunkshift are set, we can use these macros */
1701
	chunks = (blocks + CHUNK_BLOCK_RATIO(bitmap) - 1) >>
1702
			CHUNK_BLOCK_SHIFT(bitmap);
1703
	pages = (chunks + PAGE_COUNTER_RATIO - 1) / PAGE_COUNTER_RATIO;
1704 1705 1706 1707 1708 1709 1710 1711 1712 1713

	BUG_ON(!pages);

	bitmap->chunks = chunks;
	bitmap->pages = pages;
	bitmap->missing_pages = pages;
	bitmap->counter_bits = COUNTER_BITS;

	bitmap->syncchunk = ~0UL;

1714
#ifdef INJECT_FATAL_FAULT_1
1715 1716
	bitmap->bp = NULL;
#else
1717
	bitmap->bp = kzalloc(pages * sizeof(*bitmap->bp), GFP_KERNEL);
1718
#endif
1719
	err = -ENOMEM;
1720
	if (!bitmap->bp)
1721
		goto error;
1722 1723 1724

	/* now that we have some pages available, initialize the in-memory
	 * bitmap from the on-disk bitmap */
1725 1726 1727 1728 1729 1730
	start = 0;
	if (mddev->degraded == 0
	    || bitmap->events_cleared == mddev->events)
		/* no need to keep dirty bits to optimise a re-add of a missing device */
		start = mddev->recovery_cp;
	err = bitmap_init_from_disk(bitmap, start);
1731

1732
	if (err)
1733
		goto error;
1734 1735 1736 1737

	printk(KERN_INFO "created bitmap (%lu pages) for device %s\n",
		pages, bmname(bitmap));

1738 1739
	mddev->bitmap = bitmap;

1740
	mddev->thread->timeout = mddev->bitmap_info.daemon_sleep;
1741
	md_wakeup_thread(mddev->thread);
1742

1743 1744 1745
	bitmap_update_sb(bitmap);

	return (bitmap->flags & BITMAP_WRITE_ERROR) ? -EIO : 0;
1746 1747 1748 1749

 error:
	bitmap_free(bitmap);
	return err;
1750 1751
}

1752 1753 1754 1755
static ssize_t
location_show(mddev_t *mddev, char *page)
{
	ssize_t len;
1756
	if (mddev->bitmap_info.file)
1757
		len = sprintf(page, "file");
1758
	else if (mddev->bitmap_info.offset)
1759
		len = sprintf(page, "%+lld", (long long)mddev->bitmap_info.offset);
1760
	else
1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811
		len = sprintf(page, "none");
	len += sprintf(page+len, "\n");
	return len;
}

static ssize_t
location_store(mddev_t *mddev, const char *buf, size_t len)
{

	if (mddev->pers) {
		if (!mddev->pers->quiesce)
			return -EBUSY;
		if (mddev->recovery || mddev->sync_thread)
			return -EBUSY;
	}

	if (mddev->bitmap || mddev->bitmap_info.file ||
	    mddev->bitmap_info.offset) {
		/* bitmap already configured.  Only option is to clear it */
		if (strncmp(buf, "none", 4) != 0)
			return -EBUSY;
		if (mddev->pers) {
			mddev->pers->quiesce(mddev, 1);
			bitmap_destroy(mddev);
			mddev->pers->quiesce(mddev, 0);
		}
		mddev->bitmap_info.offset = 0;
		if (mddev->bitmap_info.file) {
			struct file *f = mddev->bitmap_info.file;
			mddev->bitmap_info.file = NULL;
			restore_bitmap_write_access(f);
			fput(f);
		}
	} else {
		/* No bitmap, OK to set a location */
		long long offset;
		if (strncmp(buf, "none", 4) == 0)
			/* nothing to be done */;
		else if (strncmp(buf, "file:", 5) == 0) {
			/* Not supported yet */
			return -EINVAL;
		} else {
			int rv;
			if (buf[0] == '+')
				rv = strict_strtoll(buf+1, 10, &offset);
			else
				rv = strict_strtoll(buf, 10, &offset);
			if (rv)
				return rv;
			if (offset == 0)
				return -EINVAL;
1812 1813
			if (mddev->bitmap_info.external == 0 &&
			    mddev->major_version == 0 &&
1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848
			    offset != mddev->bitmap_info.default_offset)
				return -EINVAL;
			mddev->bitmap_info.offset = offset;
			if (mddev->pers) {
				mddev->pers->quiesce(mddev, 1);
				rv = bitmap_create(mddev);
				if (rv) {
					bitmap_destroy(mddev);
					mddev->bitmap_info.offset = 0;
				}
				mddev->pers->quiesce(mddev, 0);
				if (rv)
					return rv;
			}
		}
	}
	if (!mddev->external) {
		/* Ensure new bitmap info is stored in
		 * metadata promptly.
		 */
		set_bit(MD_CHANGE_DEVS, &mddev->flags);
		md_wakeup_thread(mddev->thread);
	}
	return len;
}

static struct md_sysfs_entry bitmap_location =
__ATTR(location, S_IRUGO|S_IWUSR, location_show, location_store);

static ssize_t
timeout_show(mddev_t *mddev, char *page)
{
	ssize_t len;
	unsigned long secs = mddev->bitmap_info.daemon_sleep / HZ;
	unsigned long jifs = mddev->bitmap_info.daemon_sleep % HZ;
1849

1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941
	len = sprintf(page, "%lu", secs);
	if (jifs)
		len += sprintf(page+len, ".%03u", jiffies_to_msecs(jifs));
	len += sprintf(page+len, "\n");
	return len;
}

static ssize_t
timeout_store(mddev_t *mddev, const char *buf, size_t len)
{
	/* timeout can be set at any time */
	unsigned long timeout;
	int rv = strict_strtoul_scaled(buf, &timeout, 4);
	if (rv)
		return rv;

	/* just to make sure we don't overflow... */
	if (timeout >= LONG_MAX / HZ)
		return -EINVAL;

	timeout = timeout * HZ / 10000;

	if (timeout >= MAX_SCHEDULE_TIMEOUT)
		timeout = MAX_SCHEDULE_TIMEOUT-1;
	if (timeout < 1)
		timeout = 1;
	mddev->bitmap_info.daemon_sleep = timeout;
	if (mddev->thread) {
		/* if thread->timeout is MAX_SCHEDULE_TIMEOUT, then
		 * the bitmap is all clean and we don't need to
		 * adjust the timeout right now
		 */
		if (mddev->thread->timeout < MAX_SCHEDULE_TIMEOUT) {
			mddev->thread->timeout = timeout;
			md_wakeup_thread(mddev->thread);
		}
	}
	return len;
}

static struct md_sysfs_entry bitmap_timeout =
__ATTR(time_base, S_IRUGO|S_IWUSR, timeout_show, timeout_store);

static ssize_t
backlog_show(mddev_t *mddev, char *page)
{
	return sprintf(page, "%lu\n", mddev->bitmap_info.max_write_behind);
}

static ssize_t
backlog_store(mddev_t *mddev, const char *buf, size_t len)
{
	unsigned long backlog;
	int rv = strict_strtoul(buf, 10, &backlog);
	if (rv)
		return rv;
	if (backlog > COUNTER_MAX)
		return -EINVAL;
	mddev->bitmap_info.max_write_behind = backlog;
	return len;
}

static struct md_sysfs_entry bitmap_backlog =
__ATTR(backlog, S_IRUGO|S_IWUSR, backlog_show, backlog_store);

static ssize_t
chunksize_show(mddev_t *mddev, char *page)
{
	return sprintf(page, "%lu\n", mddev->bitmap_info.chunksize);
}

static ssize_t
chunksize_store(mddev_t *mddev, const char *buf, size_t len)
{
	/* Can only be changed when no bitmap is active */
	int rv;
	unsigned long csize;
	if (mddev->bitmap)
		return -EBUSY;
	rv = strict_strtoul(buf, 10, &csize);
	if (rv)
		return rv;
	if (csize < 512 ||
	    !is_power_of_2(csize))
		return -EINVAL;
	mddev->bitmap_info.chunksize = csize;
	return len;
}

static struct md_sysfs_entry bitmap_chunksize =
__ATTR(chunksize, S_IRUGO|S_IWUSR, chunksize_show, chunksize_store);

1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
static ssize_t metadata_show(mddev_t *mddev, char *page)
{
	return sprintf(page, "%s\n", (mddev->bitmap_info.external
				      ? "external" : "internal"));
}

static ssize_t metadata_store(mddev_t *mddev, const char *buf, size_t len)
{
	if (mddev->bitmap ||
	    mddev->bitmap_info.file ||
	    mddev->bitmap_info.offset)
		return -EBUSY;
	if (strncmp(buf, "external", 8) == 0)
		mddev->bitmap_info.external = 1;
	else if (strncmp(buf, "internal", 8) == 0)
		mddev->bitmap_info.external = 0;
	else
		return -EINVAL;
	return len;
}

static struct md_sysfs_entry bitmap_metadata =
__ATTR(metadata, S_IRUGO|S_IWUSR, metadata_show, metadata_store);

static ssize_t can_clear_show(mddev_t *mddev, char *page)
{
	int len;
	if (mddev->bitmap)
		len = sprintf(page, "%s\n", (mddev->bitmap->need_sync ?
					     "false" : "true"));
	else
		len = sprintf(page, "\n");
	return len;
}

static ssize_t can_clear_store(mddev_t *mddev, const char *buf, size_t len)
{
	if (mddev->bitmap == NULL)
		return -ENOENT;
	if (strncmp(buf, "false", 5) == 0)
		mddev->bitmap->need_sync = 1;
	else if (strncmp(buf, "true", 4) == 0) {
		if (mddev->degraded)
			return -EBUSY;
		mddev->bitmap->need_sync = 0;
	} else
		return -EINVAL;
	return len;
}

static struct md_sysfs_entry bitmap_can_clear =
__ATTR(can_clear, S_IRUGO|S_IWUSR, can_clear_show, can_clear_store);

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
static ssize_t
behind_writes_used_show(mddev_t *mddev, char *page)
{
	if (mddev->bitmap == NULL)
		return sprintf(page, "0\n");
	return sprintf(page, "%lu\n",
		       mddev->bitmap->behind_writes_used);
}

static ssize_t
behind_writes_used_reset(mddev_t *mddev, const char *buf, size_t len)
{
	if (mddev->bitmap)
		mddev->bitmap->behind_writes_used = 0;
	return len;
}

static struct md_sysfs_entry max_backlog_used =
__ATTR(max_backlog_used, S_IRUGO | S_IWUSR,
       behind_writes_used_show, behind_writes_used_reset);

2016 2017 2018 2019 2020
static struct attribute *md_bitmap_attrs[] = {
	&bitmap_location.attr,
	&bitmap_timeout.attr,
	&bitmap_backlog.attr,
	&bitmap_chunksize.attr,
2021 2022
	&bitmap_metadata.attr,
	&bitmap_can_clear.attr,
2023
	&max_backlog_used.attr,
2024 2025 2026 2027 2028 2029 2030
	NULL
};
struct attribute_group md_bitmap_group = {
	.name = "bitmap",
	.attrs = md_bitmap_attrs,
};